Ion-Molecule Reactions in the Gas Phase - ACS Publications

10 Reactions of Thermal Energy Ions by Pulsed Source Mass Spectrometry A.

G.

HARRISON

and

J.

J.

MYHER

University of Toronto, Toronto, Canada

Downloaded by FUDAN UNIV on January 30, 2017 | http://pubs.acs.org Publication Date: January 1, 1967 | doi: 10.1021/ba-1966-0058.ch010

J. C. J. THYNNE University of Edinburgh, Edinburgh, Scotland

Pulsed source techniques have been used to study thermal energy ion-molecule reactions. For most of the proton and H atom transfer reactions studied k(thermal)/k(10.5 volts/cm.) is approximately unity in apparent agreement with predictions from the simple ion-induced dipole model. However, the rate constants calculated on this basis are considerably higher than the experimental rate constants indicating reaction channels other than the atom transfer process. Thus, in some cases at least, the relationship of k(thermal) to k(10.5 volts/cm.) may be determined by the variation of the relative importance of the atom transfer process with ion energy rather than by the inter action potential between the ion and the neutral. For most of the condensation ion-molecule reactions studied k(thermal) is considerably greater than k(10.5 volts/cm.).

In mass spectrometric studies of ion-molecule reactions by conventional internal ionization techniques the reactant ions undergo continuous acceleration b y the electric field required to withdraw the ions for mass analysis. Consequently, the experimental cross-sections represent a n appropriately weighted average for ions with energies ranging from zero to the final exit energy as determined b y the electric field strength, E, and the distance of ion travel. Extensive investigations of the dependence of the cross-section on the ion velocity or field strength have been carried out, but frequently they have yielded disagreeing results, particularly at low field strengths. Stevenson and colleagues (11, 38) have found an β-in dependence for a number of simple systems i n agreement with predictions (11) from the ion-induced dipole model. O n the other hand, Field et al. (6, 8) have found an E~ dependence while H a m i l l and co workers (29) have found the energy dependence varies with the final ion exit energy. I n part these discrepancies may be caused by instrumental x

150

Ausloos; Ion-Molecule Reactions in the Gas Phase Advances in Chemistry; American Chemical Society: Washington, DC, 1967.


10.

HARRISON E T AL.

Thermal Energy Ions

151

problems such as electronic space charge and discrimination effects which can be expected to become serious at low field strengths. Some evidence supporting this has been presented recently by R y a n and Futrell {31). M o r e direct observations of the kinetic energy dependence of crosssections should be possible using external ionization techniques where the reactant ion can be chosen by initial mass analysis and, in principle, its energy more readily controlled. Several studies using external ionization techniques, both with {2, 10, 45) and without {20, 21, 27, 41) preliminary mass selection of the reactant ion, have been reported. However, apparently with these techniques it is not possible to obtain well-defined primary ion beams at energies below 0.5-1 e.v. a region of critical i m portance both experimentally and theoretically. Because of these limitations to both the conventional internal ionization and external ionization techniques it has been impossible to obtain accurate kinetic data for the reactions of thermal energy ions. Such thermal energy data are desirable for applications in such areas as radiation chemistry, discharge and combustion phenomena, and upper atmosphere research. In 1960 TaFroze and Frankevich {39) first described a pulsed mode of operation of an internal ionization source which permits the study of ion-molecule reactions at energies approaching thermal energies. In this technique a short pulse of electrons is admitted to a field-free ion source to produce the reactant ions by electron impact. A known and variable time later, a second voltage pulse is applied to withdraw the ions from the ion source for mass analysis. In the interval between the two pulses the ions react under essentially thermal conditions, and from variation of the relevant ion currents with the reaction time the thermal rate constants can be estimated. In the past few years this pulsing technique has been used by several groups, utilizing both magnetic deflection {16, 31, 37) and time-of-flight {12, 13) instruments, to study ion-molecule reactions at thermal energies. Here we review the results obtained and discuss the applications and limitations of the method, based on our observations and experiences over the past three years. Design and Operation

of Pulsed

Source

T h e construction and operating characteristics of pulsed ion sources used in conjunction with a magnetic deflection instrument have been described previously {16, 31, 37, 39); only a brief description need be given here. Figure 1 shows a schematic diagram of the source used in our laboratory in a 6-inch radius 90° sector-field instrument. T h e source is basically of the Nier design, modified by adding an electron beam control plate (plate 2) and by increasing the dimensions of the ionization chamber to reduce ion loss at the walls. In operation the electron beam is initially arrested by applying a small bias negative with respect to the filament to plate 2. T h e pulsed electron beam is then provided by applying a positive pulse to this plate.


152

ION-MOLECULE REACTIONS IN T H E GAS

1946V) 1

3

2000V+.1 1

4

2000 V

5

2000 V 2000 V + I850-2000V

(I950V1 3S Ο

6

31 7

8

PHASE

GROUND


9 10 GROUND

« · — I CM

•

Figure 1. Schematic diagram of ion source: (1) filament; (2) electron beam control plate; (3) repeller; (4) trap; (5) cage; (6) withdrawal plate; (7) focus; (8, 10) beam-defining plates; (9) beam centering plate T o prevent withdrawal of the ions thus produced by penetration of the main accelerating field, either a small positive bias is applied to plate 6 or alternatively (31) the exit slit from the ionization chamber is covered by a transparent wire mesh. T h e ions are withdrawn from the ionization chamber by a voltage pulse of proper sign applied either to the repeller plate (plate 3) or to the ion withdrawal plate (plate 6). T h e pulsed circuitry of the time-of-flight mass spectrometer is ideally suited for this type of operation, and both H a n d and von Weyssenhoff (12, 13) and L a m p e and Hess (25) have reported experiments using such an instrument with variable delay between the ionizing and withdrawal pulses to study secondary processes. In our experience the most satisfactory criterion for judging the operation of the pulsed source is the shape of the ion decay curves—i.e., the rate at which the ions are lost from the region sampled by the ion withdrawal pulse. T h i s topic is discussed in detail in the next section. Typically with our source a bias of 3-5 volts on plate 2 is sufficient to arrest the electron beam while a positive pulse of 5-10 volts amplitude and 0.1-1 μββα duration is applied to produce the pulse of electrons. T h e maximum repetition rate is about 2 Χ 10 pulses s e c . , and in practice the pulse duration is determined by the sensitivity of the detec tion system. Higher settings of the bias potential and the pulse ampli tude frequently resulted i n anomalous features in the ion decay curves, and these settings are adjusted to give the best decay curves. T o ensure removal of all ions during the withdrawal period, a pulse of at least 1 μββα duration is applied to the repeller plate. We found that the recorded ion current depends on both the duration and amplitude of these pulses, and in practice they are adjusted to give maximum ion current. These set5

-1


10.

HARRISON E TA L .

153

Thermal Energy Ions

tings as well as the focus plate voltages are mass dependent and are thus readjusted for each mass.


Ion Decay Curves and Discrimination

Effects

The usefulness of the pulsed ion source in studying ionic reactions with time as the experimental variable is determined by the rate at which the ions are lost from the volume sampled when the withdrawal pulse is applied. Ions are lost from this volume owing to their thermal kinetic energy (plus any kinetic energy involved in the process of formation) and under the influence of fields arising from surface potentials, space charge, and field penetration. All of these effects will lead to a more rapid loss of the lighter mass ion and hence will lead to discrimination effects which may obscure the changes anticipated from the reaction under study. Withdrawal

Electron

Figure 2.

Ion formation and sampling zones

The problem of ion loss can be understood more clearly by referring to Figure 2. The ions are formed initially in a volume defined by the dimensions of the electron beam, which intersects symmetrically the volume sampled by the withdrawal pulse. In our source the dimensions of the electron beam are approximately χ = 0.05 cm., y = 0.30 cm., ζ = 1.80 cm., while for the sampling volume a = 0.95 cm., β = 0.15 cm. 7 = 1.0 cm. Thus, ion loss will occur across the plane at —a/2 and by deficit in flow across the planes at ±0/2 and ±γ/2. The mathe matical formulation of the problem has been presented previously (37). Figure 3a shows a calculated decay curve for m/e = 40(Ar ) compared with two experimental curves. Curve A was obtained by pulsing the repeller plate only while curve Β was obtained by pulsing the withdrawal plate (plate 6) 50 volts negative simultaneously with the repeller pulse. In both cases the experimental curves fall off more rapidly than the theoretical curve although the shapes are the same. These results sug+



154

I O N - M O L E C U L E R E A C T I O N S I N T H EGAS P H A S E

gest that stray fields may be influencing the rate of ion loss although this more rapid decay may be caused partly by other instrumental factors. W i t h increasing delay time the fraction of the ions i n the central region of the sampling volume will tend to decrease while the fraction i n peripheral regions will tend to increase. A p p l y i n g the repeller pulse will push these peripheral ions out with a greater angle of divergence from the χ axis, and these ions will be less efficiently focussed by the ion gun. T h e effect of the withdrawal plate pulse is probably to increase the focussing ef ficiency of the ion gun. Similar results were obtained using a U-shaped repeller i n place of the normal flat repeller. T y p i c a l decay curves for various ionic masses are shown i n Figure 3b, where the effect of differing thermal velocity is clearly evident. N e g lect of such decay characteristics can lead to serious discrimination effects. Note that a l l the experimental curves have the sigmoid shape

1

I

ι

ι

ι

1

Ν.

75

=> Ο

>< Ο

1

3α

100

or or

1

CALCULATED

50

25H

Ι 100

1

1

1

1 1

1 1

1 1

1 1

1 1

1 1

1

1

3b

(Τ LU Ν

75r-

50 —

25

0

1

1

2

1

1

3

1

4

1

5

1

6

7

8

DELAY TIME ( μ sec.) Figure 3. (a) Comparison of experimental and calculated decay curves; m/e = 40. (b) Experimental decay curves for various masses



10.

155

Thermal Energy Ions

HARRISON E T A L .

predicted by theory and that the total ion loss in the first part of the delay period is small. Therefore, we have restricted our observations to this time range and, further, have required that the decay curves have this characteristic shape before considering the results reliable. In addition to the effects discussed above, two further possible sources of discrimination peculiar to ion-molecule reactions must be considered. First, although it is known that most primary ions are formed without kinetic energy, such may not be the case for ions pro duced by ion molecule reactions. Secondary ions formed in exothermic ion-molecule reactions could retain a considerable fraction of the exothermicity as kinetic energy and diffuse from the sampling region at a rate considerably greater than predicted from the ambient temperature. T h e limited evidence to date (40) indicates that the kinetic energy of the product ions is small, but this may not be true for all types of reactions. Second, the primary and secondary ions are not formed in the same region and this may lead to different collection efficiencies. In the ex treme, primary ions initially formed in the sampling region may diffuse out of the sampling region during the delay period (32, 39). If these ions react after leaving the sampling region, the product ion may re-enter the sampling region (assuming the formation process imparts no preferential direction of travel), thus leading to anomalously high secondary ion con centrations. T h i s effect will be serious only i f an appreciable fraction of the reactant ions is lost from the sampling region during the delay pe riod. Identification

of Reactant

and Product

Ions

T h e variation of ion intensities with reaction time should prove to be a useful approach in identifying reactant and product ions in ion-molecule reactions. T h e approach has been used little, primarily because most studies have been carried out at relatively low source concentrations where the extent of reaction (and therefore the changes in relative i n tensities) is small. Some indication of the potential usefulness is shown by ion-molecule reactions in dimethyl ether (15). Figure 4 shows the variation of J ( ( C H ) O H ) + , J ((CH ),0)+, and 7 ( C H O C H ) + as a function of the delay time at an electron energy where only ( C H ) 0 and C H 3 O C H 2 are significant primary ions. T h e results are expressed as a percentage of the total ionization to allow for ion decay. T h e increase in / 4 7 / Σ / clearly shows its secondary character while the decrease in IΆβ/Σΐ shows that ( C H ) 0 + is a reactive ion. T h e constancy of 145/Σι indicates that C H O C H is nonreactive under these conditions. Similar studies have been carried out (43) on the ammonia system in an attempt to detect the reaction: 4 7

3

2

4 6

S

4 5

3

2

3

2

+

+

3

3

2

2

+

N H + + N H -> N H + + N H 2

3

3

2

(1)

T h e intensity changes observed, while consistent with the occurrence of Reaction 1, were small enough to have originated from discrimination effects. A new instrument is being constructed in our laboratory which



156

ION-MOLECULE REACTIONS IN T H E GAS

DELAY TIME Figure 4.

PHASE

(μεβα)

Ion intensities as a function of delay time, CHtOCHs

should permit operation at considerably higher pressures, thus making the changes caused by chemical reaction much greater compared with ion decay. Charge

Transfer

Reactions

It is particularly difficult to study charge transfer reactions by the usual internal ionization method since the secondary ions produced will always coincide with ions produced i n primary ionization processes. Indeed these primary ions frequently constitute the major fraction of the total ion current, and the small intensity changes originating from charge transfer reactions are difficult to detect. F o r example, Field and F r a n k lin (5) were unable to detect any charge transfer between X e and C H by the internal ionization method although such reactions have been observed using other techniques (3, 9,22). +


4

10.

157

Thermal Energy Ions

HARRISON E T AL.

T h e pulsing mode of operation of the internal ionization source should afford decided advantages for studying charge transfer reactions and should permit one to evaluate the rates at thermal energies. F o r the charge transfer reaction A + Β Β + A , one should be able to observe increases in the ratio Β / A + with increasing reaction time. However Β and A will often differ considerably in mass, and discrimination effects may be much greater than changes owing to charge transfer. In the only study to date of such reactions by the pulsing method Karachevtsev et al. (24) have utilized this mass discrimination effect to advantage in studying the reactions of the rare gas ions A r , K r +, X e with C H , C H , and C H . T h e y deliberately used a long delay (14-20 /xsec.) between the ionizing and withdrawal pulses. During this delay period, the lighter mass (and thus higher velocity) hydrocarbon ions formed in the initial impact process are lost preferentially from the sam pling region. Hydrocarbon ions formed by charge transfer from the heavier rare gas ions, especially those formed during the latter stages of the delay period, will not be lost, and the hydrocarbon spectrum obtained when the withdrawal pulse is applied shows a considerable enrichment in secondary hydrocarbon ions compared with primary ions. F r o m the variation of this spectrum with rare gas pressure the relative rates of formation of the hydrocarbon fragments were estimated. Table I com pares the charge transfer spectrum obtained i n this way for A r + C H with spectra obtained by other techniques. T h e agreement is quite good. +

+

+

+


4

2

4

2

+

6

+

4

B y adding small amounts of H to the gas mixture and observing the rate of formation of A r H , they also estimated the following overall rate constants for charge transfer at thermal ion energies. 2

+

Ar+ + C H

k = 7.5 X 10 " cc. molecule" 9

4

Ar+ + C H

4

* = 3.9 Χ 1 0 "

Ar+ + C H

6

k = 6.3 X 10~

2

2

1

sec."

1

9

9

These rate constants were measured relative to the rate constant for the reaction: Ar+ + H

2

ArH

+

+ H

(2)

which was taken to have a rate constant of 1.68 X 10~ cc. m o l e c u l e sec. (38). Note however, that the rate constant for reaction 2 was not 9

-1

_ 1

Table I. Ar

+

Ionic Abundances from Charge Transfer of A r w i t h C H

Energy e.v.

Thermal 0-0.2 0-2.5 25 25 75 500

+

CH + 4

19 " 1 ' 2.3 2 4

Relative Abundance CH S

100 100 100 100 100 100 100

+

CH

2

+

Reference

21 19 24 24 24 32.8 23


24 28 7 9 3 23 26

4

158

I O N - M O L E C U L E R E A C T I O N S IN T H E GAS P H A S E

measured at thermal energies and further neglects the contribution of the concurrent reaction: H +

+ Ar-> ArH+ + H

2

(3)

which has been shown to occur (35). T h e overall rate constant for the A r - C H system is appreciably higher than earlier values of 1.1 X 10 ~ (7) and 1.9 X 10~~ cc. m o l e c u l e s e c . (28) obtained using conventional internal ionization techniques. 9

4

9


Proton

-1

and Hydrogen

Atom

_ 1

Transfer

Reactions

Reactions involving a transfer of a proton or a hydrogen atom are an extremely common type of ion-molecule reaction and are particularly suited for study by the pulsed source technique. T h e secondary ion will usually occur at an m je ratio where it is not obscured by abundant pri mary ions, and the product and reactant ions frequently will differ only slightly in mass, thus minimizing discrimination effects. F o r the general ion-molecule reaction: P

+ M

+

S

+

+ Ν

(4)

studied with reaction time as the experimental variable and assuming the neutral concentration [M ] to be constant, the integrated rate expression becomes ln

(|p-t) * =

[M]i

(5)

where [P ] and [P + represent the concentration of Ρ timet Writing for [P ] +

0

+

+

at time zero and

0

[P h +

[P ]o =

+

+

l$ h

(6)

+

yields ln(l +

(

|^)

=A[MJi

(7)

which for [S + ] / [P ] < < 1 can be approximated by: +

[ S ] , / [ P ], = k[M]t +

(8)

+

In the pulsed system, reaction will take place not only during the delay period but also during the time when the ionizing and withdrawing pulses are being applied. F o r constant pulse durations and amplitudes the secondary ions produced during these periods can be represented by a constant term added to Equations 7 and 8. Assuming further that the collection efficiencies are the same for both primary and secondary ions, this leads to following equations in terms of the measurable ion currents ln ^1 + | ^

= k[M]t

+

C


(9)

10.

HARRISON E T A L .

Thermal Energy Ions

159

—

(10)

Ip +

= k [M]f + C

1

where C and C are constants reflecting the reaction occurring during the ionizing and withdrawal pulses, and t represents the delay time between the ionizing and withdrawal pulses. Figure 5 shows the experimental results for several reactions plotted in the form of Equation 10. In all cases satisfactory straight lines are found which agree with predictions. T h e positive intercepts represent the contributions owing to reactions during the ionizing and withdrawal pulses plus, in some cases, isotope contributions at the m/e ratio of the secondary ion. In several cases an upward curvature is discernible at longer delay times, and this is u n doubtedly caused by mass discrimination effects. Downloaded by FUDAN UNIV on January 30, 2017 | http://pubs.acs.org Publication Date: January 1, 1967 | doi: 10.1021/ba-1966-0058.ch010

1

2

6h-

σ4

σβ

ο·6

ΙΌ

DELAY TIME ( μ sec.) Figure 5.

Typical Ι & / Ι Ρ +

+

VS. delay time plots

In several systems studied the secondary ion is formed by more than one ion-molecule reaction. F o r example, we have found (43) that the NH ion is formed in the ammonia system by the concurrent reactions: 4

+

NH + + N H 3

3

NH

+

4

+ NH

NH + + N H , • •NH + + 2

4

2

(ID

NH

(12)


160

ION-MOLECULE REACTIONS

IN T H E GAS PHASE

For the two concurrent reactions with low percent reaction the total secondary ion current 7NH after delay time t can be expressed as the sum of two independent contributions: 4

i W

=

+

[£II/NH3 [NH ] +

3

+

^i2W[NH ]]i 3

+

D

(13)

1

(14)

Rearrangement leads to: *n[NH ]


8

+

t +

*i [NH,]2

D

Equation (14) predicts that for constant 7 Ν Π / / Ν Π the plots of 7ΝΠ / / N H S VS. delay time should be linear and with slopes proportional to /NH2+//NH3+. Figure 6 shows typical plots of / N H 2 / ^ N H VS. delay time t while Figure 7 shows the slopes of such delay time plots as a function of the ratio 2 Ν Η / / Ν Η , this ratio being changed by variation of the elec tron energy. A good straight line results in agreement with predictions. Using either of the above approaches we have measured the thermal rate constants for some 40 hydrogen atom and proton transfer reactions. T h e results are tabulated i n Table II where the thermal rate constants are compared with the rate constants obtained at 10.5 volt c m . (3.7 e.v. exit energy) either by the usual method of pressure variation or for concurrent reactions by the ratio-plot technique outlined i n previous publications (14, 17, 36). T h e ion source temperature during these measurements was about 3 1 0 ° K . Table II also includes the thermal rate constants measured by others (12, 13, 33, 39) using similar pulsing tech niques. 2

+

3

+

4

+

+

2

+

3

3

+

+

- 1


+


T a b l e II.

Proton and H A t o m Transfer Ion-Molecule Rate Constant Χ ΙΟ cc. molecule' sec.' 10.5 Thermal volts/cm.

Reactions

10

1

Reaction

H+ + H -*H ++ H H D + + H D -> H D + + D H D + H D -*» H D + H D + D D + D NH + + N H NH + NH N H + + N H — NH + + N H OH+ + H 0 H 0 + Ο H 0+ + H 0 H 0 + OH 2

2

3

2

+

2

+

2

+

2

+

3

2

3

3

3

4

2

2

+

4

2

2

+

3

2

+

3

8

OD + + D 0 -+ D 0 + + Ο D 0+ + D 0 D 0 + + OD N H + + N H N H + N H N H + + N H -> N H + + N H HS + H S - H S + S H S + H S — H S+ + HS C H + + C H -> CH + + C H 2

2

3

2

3

2

4

2

2

4

2

4

2

+

2

+

2

5

+

5

4

5

3

3

4

3

3

+

2

3

2

3

2

2

2

3

2

2

2

2

2

3

2

+

3

2

3

3

3

3

4

+

2

4

4

+

4

5

3

3

+

2

8

7

k {thermal) k(10.5 volts/cm.) Reference

13 .3 8 .4 6 .9 15 .0 5 1 7 .7 5 9i 6 0

0 44 0 .45 0 42 0 38 0 67 0 68 0 86 0 82

2 6 3 .5 3 8 5 5 2 6 2 1 7 5 10 7

0 1 0 0

6

0 76

15 15 15 15 44 44 43 43 39 43 43 44 44 18 18 15 33 39 12 15

3

0 75

15

4

4

4 3 2

8

5 8

85 06 27 25 ]L 3 ]L.3 0 81 0 90

6 3

+

1 3o

1 7

+ +

2 7

4 Is

0 65

15

2 le

3 3

0 65

15

1 6

4

2 3o

0 71

15

3 3

6

4 9o

0 69

15

6. 3 5. 2

0 74 0. 38

15 15

2

3

2 3

3

C H D + + C H D -> C H D + CH D C H D + + C H D -> C H D + CH CH D + + CH D — CH D CHD CH D + + CH D — CH D CH D CHD + CHD - C H D CD C H D + + C H D -> C H D CHD CD + C D — C D + GD GD + + GO -> CDO+ + GD 2

2

3

4

e

2

+

3

2

2

5

3

2

5 .9 3 .8 2 .9 5 .8 3 .4 5 .2 4 .7 4 .9 8 .5" 2 2 3 .6 1 0 1 3 3 6 2 8 6 1·> 9 6 11 0 11 6 4 4

1

+

+

+ +

0

4. 7i 1. %

6

5 2


162

I O N - M O L E C U L E R E A C T I O N S IN T H E GAS P H A S E

T a b l e II.

Continued

Rate Constant Χ 10 cc. molecule~ sec.' 10.5 Thermal volts/cm.

10

l

Reaction CO+ + D — CDO+ + D CO + C D -> C D O + + C D C H + C H — C H + + C H C H O H + + C H 3 O H -> CH OH + + ? CH OH + C H 3 O H -> CH OH + + CH 0 CD OH + + CD OH CD OH + + CD 0 CD OH + + CD OH CD OHD + + CD OH CD OH + + CD OH CD OH + + CD 0 CD OH+ + CD OH — CD OHD+ + CDOH (CH ) 0 + (CH ) 0 (CH ) OH+ + C H O C H CH OCD + + CH OCD — CH OCD H + + CH OCD CH OCD + + CH OCD — C H O C D D + + CH3OCD2 CH CHO+ + CH CHO CH CHOH + ? C H O + + C H C H O -> CH CHOH+ + CO CD + CH CHO — CH CHOD + + CD CH CN + CH CN — CH CNH+ + CH CN CH C1+ + CH C1 — CH C1H+ + CH C1 CH C1+ + CH C1 -H. CH C1H+ + CHC1

7. .16* 2. 6 7 3. 5

2

+

2

4

3

+

3

2

4

2

5

2

2

0. 48 1. 0

13 15 30

0. .84

42

5

]L . l

42

6 . .3o

0. .79

42

0. .73

42

1. 0

42

5. • 5 3. .60

8

k(thermal) k\Ï0~5 volts/cm.) Reference

1

7

3

3

2

3

2

14. .6

9. .0

8. 3

2

3

3

3

2

3

3

4. 9

7

5. 9

9

8. • 2

7. l

a

6 •6

3

3


12. 2

+

2

2

2

6

3

3

2

2

2

8

8

3

3

3

+

2

3

3

2

3

3

3

3

3

9. 4

9

8. • 2

4

1.

5. 3

6

4•7

6

1. • Is

15

3

4. 3

9

1. 0

15

15

I5

3

3

3

3

42

6

3

2

3

0 .4

2

3

3

3

Measured value neglecting contribution of reaction D 2 + C O COD + D. +

+

+

+

A s Table II shows, four separate measurements of the thermal rate constants for the reaction: CH + + C H 4

4

CH + + C H 5

(15)

3

have been made, yielding rate constants between 6.1 X 10 ~ and 11.6 X 10 ~ cc. molecule s e c . T h i s spread is probably a reasonably true representation of the absolute accuracy of the technique at the present time. One of the major sources of error is undoubtedly the estimation of the source concentration of neutral molecules. Relative values obtained in one laboratory should be somewhat more accurate. F o r ion-molecule reactions where the interaction can be attributed to ion-induced dipole forces it has been shown (11) that the rate constant should be independent of ion energy—i.e., the thermal and 10.5 volt c m . rate constants should be the same. T h e third column in Table II shows that for most of the reactions studied the ratio k (thermal)/& (10.5 volts/ cm.) is in the range 0.7-1.1. Considering the errors involved this is not significantly different from unity, indicating that most of the reactions 10

10

_ 1

_ 1

_ 1


10.

163

Thermal Energy Ions

HARRISON E T A L .

agree approximately with the ion-induced dipole model in this energy range. T h i s appears to include a number of reactions where one might expect ion-permanent dipole interactions to be significant. A few note worthy cases where the ratio of rate constants is not unity are the H , H D , D , C O - C D 4 , and N H systems where k(thermal)(10.5 volts/cm.) X H + + X 2

(17)

where the alternative charge exchange reaction is a resonance process; indeed, Henglein and M u c c i n i (21) have observed charge transfer in addi tion to Reaction 17 for many of the systems included in Table III.


164

I O N - M O L E C U L E REACTIONS IN T H E GAS P H A S E

Since it thus appears that reactions other than the atom transfer process are occurring, one must consider the possibility that the low k(thermal)(10.5 volts/cm.) ratios may result from a variation of the relative importance of the atom transfer reaction channel with ion energy. Similarly, in some of the cases where & (thermal) = £(10.5 volts/cm.) the relative importance of the atom transfer process may also change with ion energy. T h u s the value of k(thermal)(10.5 volts/cm.) does not necessarily provide conclusive evidence for the interaction potential between the ion and the neutral molecules.


Condensation

Ion-Molecule

Reactions

Condensation reactions are somewhat more difficult to study by the pulsing technique since the secondary ion usually has a considerably higher mass than the primary ion. However, by restricting the total reaction time to 1 μββα or less, we have found it possible to study this type of reaction under thermal conditions. Preliminary results are pre sented in Table I V . M a n y of the product ions in these systems can be formed i n more than one reaction, and the details of reaction identifica tion will be presented elsewhere. Table IV.

Ion-Molecule Reactions i n Hydrocarbons Rate Constant Χ 10 cc. molecule' sec. 10.5 volts/cm. Thermal

10

k (thermal) k (10.5 volts/cm.)

Reference

6.0 4.6 0.91e 5.0 2

1.4 1.3 1.79 1.05

34 34 30 30

1

Reaction C H + + C H -* C H + H CD + CD -*C D + + D C H + + CH — C H + + Η C H + + C H -> C H + + Η C H -f" C H —*• C H -jH C H + + C H (aliène) — C H -f- C H C H + + C H (aliène) C H + + H C H + C H (propyne) C Hs + C H C D + C D (propyne) — C D + + D 3

4

2

5

+

8.6 6.1 1.6 5.2

2

+

3

2

2

2

2

2

4

2

+

2

2

5

2

4

3

5

2

4

3

2

2

3

4

4

2

2

4

3

3

+

2

4

6

3

4

3

4

7

+

4

+

3

3

3

3

4

6

+

3

3

4 6

4

8

+

2

3

1

2.2

B

I.87

1.20

30

5.4

8

3.2

3

1.70

30

3.7

3

1.4

8

2.52

30

0.91

30

1.44

30

2.2

2.0

3

4

4.7

7

3.2

3

9

For most of the reactions investigated the ratio k (thermal)(10.5 volts/cm.) is considerably above unity. One exception listed in Table I V is the reaction:

C H + + C H — C H + + CaHa 3

4

3

4

3

(18)

5

where the ratio is 0.91—i.e., in the same range as found for the simple hydrogen transfer reactions (Table II). Using mixtures of C H and C D we have found no evidence for isotopic mixing in the product ion, thus indicating that an intimate collision complex is not formed. B y contrast, reactions of C H + C H producing C and C ions occur with essentially complete scrambling of the hydrogen atoms. 3

3

4

3

4

+

3

4

5

6


4

10.

HARRISON E T A L .

Thermal Energy Ions

165

T h e results in Table I V suggest that the condensation reactions cannot be described adequately by the ion-induced dipole model. In this regard the results agree with conventional studies which have fre quently found a higher power inverse dependence of the cross-section on the field strength Ε for condensation reactions than for hydrogen transfer reactions.


Conclusions T h e pulsed source method, despite several limitations, appears to be a very useful technique for studying ion-molecule reactions at thermal energies. Although the studies to have date been limited primarily to simple hydrogen transfer reactions, the technique should also prove useful for studying charge transfer and hydride ion transfer reactions at thermal energies. In conclusion we would note the modified pulsing technique recently developed by Futrell and colleagues (32, 33) for studying ion-molecule reactions at low ion energies. T h i s approach, which is an outgrowth of the pulsing techniques described here, represents a potentially powerful tool for studying ion-molecule reactions not only at thermal energies but also over a range of ion energies not easily accessible by conventional techniques.

Literature

Cited

(1) A m e n u - K p o d o , F. K., M a t e r ' s Thesis, University of T o r o n t o , 1965. (2) B e r t a , Μ . Α . , K o s k i , W . S., J. Am. Chem. Soc. 86, 5098 (1964). (3) C e r m a k , V., H e r m a n , Z . , Nucleonics 19, 106 (1961). (4) Eliason, Μ . Α . , Hirschfelder, J. O . , J. Chem. Phys. 30, 1426 (1959). (5) F i e l d , F. H., F r a n k l i n , J. L., J. Am. Chem. Soc. 83, 4509 (1961). (6) F i e l d , F. H., F r a n k l i n , J. L., L a m p e , F. W., J. Am. Chem. Soc. 79, 2419, 2665 (1957). (7) F i e l d , F. H., H e a d , H. N., F r a n k l i n , J. L., J. Am. Chem. Soc. 84, 1118 (1962). (8) F i e l d , F. H., L a m p e , F. W., J. Am. Chem. Soc. 80, 5583 (1958). (9) G a l l i , Α . , G i a r d i n i - G u i d o n i , Α . , V o l p i , G. G., Nuovo Cimento 31, 1145 (1964). (10) Giese, C . F., M a i e r , W . B., J. Chem. Phys. 35, 1913 (1961); 39, 739 (1963). (11) Gioumousis, G., Stevenson, D . P . , J. Chem. Phys. 29, 294 (1958). (12) H a n d , C . W . , v o n Weyssenhoff, H., Can. J. Chem. 42, 195 (1964). (13) H a n d , C . W . , v o n Weyssenhoff, H., Can. J. Chem. 42, 2385 (1964). (14) H a r r i s o n , A. G., Can. J. Chem. 41, 236 (1963). (15) Harrison, A. G., Ivko, Α . , Shannon, T. W . , Can. J. Chem., 44, 1351 (1966). (16) Harrison, A. G., Shannon, T. W . , M e y e r , F., Advan. Mass Spectrometry 3, 377 (1966). (17) Harrison, A. G., T a i t , J. M. S., Can. J. Chem. 40, 1936 (1962). (18) Harrison, A. G., T h y n n e , J. C. J., Trans. Faraday Soc., i n press. (19) H a r r i s o n , A. G., T h y n n e , J. C. J., unpublished results. (20) Henglein, Α . , M u c c i n i , G., Z. Naturforsch. 17a, 452 (1962). (21) Henglein, Α . , M u c c i n i , G., Z. Naturforsch. 18a, 1753 (1963). (22) Hertel, G. R . , K o s k i , W . S., J. Am. Chem. Soc., 87, 1686 (1965). (23) Homer, J. B., Lehrle, R . S., R o b b , J. C . , T h o m a s , D . W . , Advan. Mass Spectrometry 2, 503 (1963). (24) Karachevtsev, G . V., M a r k i n , M. I., T a l ' r o z e , V. L., Kinetika i Kataliz, 5, 377 (1964). (25) L a m p e , F. W . , Hess, G. G., J. Am. Chem. Soc. 86, 2952 (1964). (26) L i n d h o l m , Ε . , Z. Naturforsch. 9 a , 535 (1954).



166

ION-MOLECULE

REACTIONS

IN THE GAS PHASE

(27) M a r k i n , M. I., T a l ' r o z e , V . L., Elementarnye Protsessy Khim. Vysokikh Energ. Akad. Νauk SSR, Inst. Khim. Fiz., Tr. Simpoziuma, Moscow, 1963, 18, (1965). (28) M e l t o n , C . E., J. Chem. Phys. 33, 647 (1960). (29) M o r a n , T. F., H a m i l l , W . H., J. Chem. Phys. 39, 1413 (1963); see also earlier references cited therein. (30) M y h e r , J. J., Harrison, A. G., unpublished results. (31) R y a n , K. R . , F u t r e l l , J. H., J. Chem. Phys. 42, 824 (1965). (32) R y a n , K. R . , F u t r e l l , J. H., J. Chem. Phys. 43, 3009 (1965). (33) R y a n , K. R . , F u t r e l l , J. H., M i l l e r , C . D., Rev. Sci. Instr. 37, 107 (1966). (34) Shannon, T. W . , Ph.D. Thesis, University of T o r o n t o (1965). (35) Shannon, T. W . , H a r r i s o n , A. G., J. Chem. Phys. 43, 4201 (1965). (36) Shannon, T. W . , H a r r i s o n , A. G., J. Chem. Phys. 43, 4206 (1965). (37) Shannon, T. W . , M e y e r , F., H a r r i s o n , A. G., Can. J. Chem. 43, 159 (1965). (38) Stevenson, D . P . , Schissler, D. O . , J. Chem. Phys. 29, 287 (1958). (39) Tal'roze, V. L., F r a n k e v i c h , E. L., Zh. Fiz. Khim. 34, 2709 (1960). (40) Tal'roze, V. L., F r a n k e v i c h , E. L., Tr. Soveshch. po Radiacionnoi Khim., Moscow, 1957, 13. (41) T a l ' r o z e , V . L., Karachevtsev, G . V., Advan. Mass Spectrometry 3, 211 (1966). (42) T h y n n e , J. C. J., A m e n u - K p o d o , F. K., Harrison, A. G., Can. J. Chem., i n press. (43) T h y n n e , J. C . J., H a r r i s o n , A. G., Trans. Fraday Soc., i n press. (44) T h y n n e , J. C. J., Harrison, A. G., unpublished results. (45) Weiner, E. R . , Hertel, G. R . K o s k i , W. S., J. Am. Chem. Soc. 86, 788 (1964). RECEIVED

M a y 2,

1966.


Ion-Molecule Reactions in the Gas Phase - ACS Publications

Recommend Documents